In [3]:
import pandas as pd
from IPython.display import Image
import numpy as np
from matplotlib import pyplot as plt
%matplotlib inline
import seaborn as sns
import plotly
import plotly.offline as py
import plotly.graph_objs as go
import plotly.io as pio
from collections import defaultdict
import collections
import os
import psutil
import warnings
warnings.filterwarnings('ignore')

reading the datasets¶

In [5]:
deli=pd.read_csv(r'deliveries.csv')
mat=pd.read_csv(r'matches.csv')

data pre processing¶

In [6]:
##same team
deli=deli.replace(to_replace="Rising Pune Supergiant",value="Rising Pune Supergiants")
mat=mat.replace(to_replace="Rising Pune Supergiant",value="Rising Pune Supergiants")

##delhi daredevils to delhi capitals
deli=deli.replace(to_replace="Delhi Daredevils",value="Delhi Capitals")
mat=mat.replace(to_replace="Delhi Daredevils",value="Delhi Capitals")

##deccan chargers to sunrisers Hyderabad
deli=deli.replace(to_replace="deccan chargers",value="sunrisers Hyderabad")
mat=mat.replace(to_replace="deccan chargers",value="sunrisers Hyderabad")

##same name of stadiums
mat=mat.replace(to_replace="MA Chidambaran Stadium",value="M. A.Chidambaram Stadium")
mat=mat.replace(to_replace="Punjab Cricket Association IS Bindra Stadium,Mohali",value="Punjab Cricket Association Stadium,Mohali")
mat=mat.replace(to_replace="Rajiv Gandhi International Stadium,Uppal",value="Rajiv Gandhi Intl.Cricket Stadium")
mat=mat.replace(to_replace="ACA-VDCA Stadium",value="Dr. Y.S. Rajeskhara reddy ACA-VDCA Stadium")
mat=mat.replace(to_replace="M. Chinnaswamy Stadium",value="M Chinnaswamy Stadium")

mat['winner']=mat['winner'].fillna("No Result")
In [7]:
deli['bowling_team'].unique()
Out[7]:
array(['Royal Challengers Bangalore', 'Sunrisers Hyderabad',
       'Rising Pune Supergiants', 'Mumbai Indians',
       'Kolkata Knight Riders', 'Gujarat Lions', 'Kings XI Punjab',
       'Delhi Capitals', 'Chennai Super Kings', 'Rajasthan Royals',
       'Deccan Chargers', 'Kochi Tuskers Kerala', 'Pune Warriors'],
      dtype=object)
In [8]:
deli['batting_team'].unique()
Out[8]:
array(['Sunrisers Hyderabad', 'Royal Challengers Bangalore',
       'Mumbai Indians', 'Rising Pune Supergiants', 'Gujarat Lions',
       'Kolkata Knight Riders', 'Kings XI Punjab', 'Delhi Capitals',
       'Chennai Super Kings', 'Rajasthan Royals', 'Deccan Chargers',
       'Kochi Tuskers Kerala', 'Pune Warriors'], dtype=object)
In [9]:
Teams={'Sunrisers Hyderabad': 'SRH', 'Royal Challengers Bangalore':'RCB',
       'Mumbai Indians': 'MI', 'Rising Pune Supergiants': 'RPS', 'Gujarat Lions': 'GL',
       'Kolkata Knight Riders': 'KKR', 'Kings XI Punjab': 'KXIP', 'Delhi Capitals': 'DD',
       'Chennai Super Kings': 'CSK', 'Rajasthan Royals': 'RR', 'Deccan Chargers':'DC',
       'Kochi Tuskers Kerala': 'KTK', 'Pune Warriors': 'PW'
       }
deli['batting_team']=deli['batting_team'].map(Teams)
deli['bowling_team']=deli['bowling_team'].map(Teams)
In [10]:
deli
Out[10]:
match_id inning batting_team bowling_team over ball batsman non_striker bowler is_super_over ... bye_runs legbye_runs noball_runs penalty_runs batsman_runs extra_runs total_runs player_dismissed dismissal_kind fielder
0 1 1 SRH RCB 1 1 DA Warner S Dhawan TS Mills 0 ... 0 0 0 0 0 0 0 NaN NaN NaN
1 1 1 SRH RCB 1 2 DA Warner S Dhawan TS Mills 0 ... 0 0 0 0 0 0 0 NaN NaN NaN
2 1 1 SRH RCB 1 3 DA Warner S Dhawan TS Mills 0 ... 0 0 0 0 4 0 4 NaN NaN NaN
3 1 1 SRH RCB 1 4 DA Warner S Dhawan TS Mills 0 ... 0 0 0 0 0 0 0 NaN NaN NaN
4 1 1 SRH RCB 1 5 DA Warner S Dhawan TS Mills 0 ... 0 0 0 0 0 2 2 NaN NaN NaN
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
179073 11415 2 CSK MI 20 2 RA Jadeja SR Watson SL Malinga 0 ... 0 0 0 0 1 0 1 NaN NaN NaN
179074 11415 2 CSK MI 20 3 SR Watson RA Jadeja SL Malinga 0 ... 0 0 0 0 2 0 2 NaN NaN NaN
179075 11415 2 CSK MI 20 4 SR Watson RA Jadeja SL Malinga 0 ... 0 0 0 0 1 0 1 SR Watson run out KH Pandya
179076 11415 2 CSK MI 20 5 SN Thakur RA Jadeja SL Malinga 0 ... 0 0 0 0 2 0 2 NaN NaN NaN
179077 11415 2 CSK MI 20 6 SN Thakur RA Jadeja SL Malinga 0 ... 0 0 0 0 0 0 0 SN Thakur lbw NaN

179078 rows × 21 columns

most successfull team¶

In [12]:
#asking user whether to save the plot on the disk or not
if(input("Save/download? y/n: ").lower()[0]=='y'):
   save_files=True
   print("\nplots will be downloaded/saved.")
else:
   save_files=False
   print("\nplots will not be downloaded/saved.")    

Save/download? y/n: n

plots will not be downloaded/saved.
In [13]:
mat['winner'].value_counts()
Out[13]:
Mumbai Indians                 109
Chennai Super Kings            100
Kolkata Knight Riders           92
Royal Challengers Bangalore     84
Kings XI Punjab                 82
Delhi Capitals                  77
Rajasthan Royals                75
Sunrisers Hyderabad             58
Deccan Chargers                 29
Rising Pune Supergiants         15
Gujarat Lions                   13
Pune Warriors                   12
Kochi Tuskers Kerala             6
No Result                        4
Name: winner, dtype: int64

teams that have won most finals¶

In [20]:
finals=mat.drop_duplicates(subset=['season'],keep='last')
finals=finals[['id','season','city','team1','team2','toss_winner','toss_decision','winner']]

most_finals=pd.concat([finals['team1'],finals['team2']]).value_counts().reset_index()
most_finals.rename({'index':'team',0:'count'},axis=1,inplace=True)
wins=finals['winner'].value_counts().reset_index()
most_finals=most_finals.merge(wins,left_on='team',right_on='index',how='outer')
most_finals=most_finals.replace(np.NaN,0)
most_finals.drop('index',axis=1,inplace=True)
most_finals.rename({'count':'finals_played','winner':'won_count'},inplace=True,axis=1)
most_finals
Out[20]:
team finals_played won_count
0 Chennai Super Kings 8 3.0
1 Mumbai Indians 5 4.0
2 Royal Challengers Bangalore 3 0.0
3 Sunrisers Hyderabad 2 1.0
4 Kolkata Knight Riders 2 2.0
5 Deccan Chargers 1 1.0
6 Kings XI Punjab 1 0.0
7 Rising Pune Supergiants 1 0.0
8 Rajasthan Royals 1 1.0
In [22]:
scores=deli.groupby(['match_id','inning','batting_team','bowling_team'])['total_runs'].sum().reset_index()
scores
Out[22]:
match_id inning batting_team bowling_team total_runs
0 1 1 SRH RCB 207
1 1 2 RCB SRH 172
2 2 1 MI RPS 184
3 2 2 RPS MI 187
4 3 1 GL KKR 183
... ... ... ... ... ...
1523 11413 2 DD SRH 170
1524 11414 1 DD CSK 155
1525 11414 2 CSK DD 162
1526 11415 1 MI CSK 152
1527 11415 2 CSK MI 157

1528 rows × 5 columns

In [24]:
scores_1=scores[scores['total_runs']>=200]
scores_1
Out[24]:
match_id inning batting_team bowling_team total_runs
0 1 1 SRH RCB 207
16 9 1 DD RPS 205
38 20 1 RCB GL 213
62 32 1 SRH KXIP 207
72 36 1 SRH KKR 209
... ... ... ... ... ...
1475 11326 2 KKR RCB 219
1488 11333 1 RCB KXIP 213
1498 11338 1 KKR MI 241
1499 11338 2 MI KKR 204
1500 11339 1 SRH KXIP 232

105 rows × 5 columns

In [27]:
fig=plt.figure(figsize=(12,8))
plt.title("teams that have scores 200+",fontsize=30)
sns.countplot(x='batting_team',data=scores_1)

if save_files:
    if not os.path.exists('plots'):
     os.makedirs('plots')
    filename='plots/teams_200s'
    plt.savefig(filename,bbox_inches='tight')

highest score in each season¶

In [30]:
matches_list=[]
total_innings=[]
batting_team=[]
runs=[]

for match_no in deli['match_id'].unique():
    for innings in deli[deli['match_id']==match_no]['inning'].unique():
        ad=deli[
                (deli['match_id']==match_no)&
                (deli['inning']==innings)
        ]
        total_runs=ad['total_runs'].sum()
        runs.append(total_runs)
        matches_list.append(match_no)
        total_innings.append(innings)
        batting_team.append(ad['batting_team'].unique()[0])
In [32]:
ad0=pd.DataFrame()
ad0['match_id']=matches_list
ad0['total_runs']=runs
ad0['season']=[mat[mat['id']==i]['season'].unique()[0] for i in matches_list]
ad0['batting_team']=batting_team
ad0
Out[32]:
match_id total_runs season batting_team
0 1 207 2017 SRH
1 1 172 2017 RCB
2 2 184 2017 MI
3 2 187 2017 RPS
4 3 183 2017 GL
... ... ... ... ...
1523 11413 170 2019 DD
1524 11414 155 2019 DD
1525 11414 162 2019 CSK
1526 11415 152 2019 MI
1527 11415 157 2019 CSK

1528 rows × 4 columns

In [41]:
season=[]
teamid=[]
max_runs=[]
for year in ad0['season'].unique():
    maximum_run=ad0[ad0['season']==year]['total_runs'].max()
    team=ad0[(ad0['season']==year)& (ad0['total_runs']==maximum_run)]['batting_team'].unique()[0]
    season.append(year)
    teamid.append(team)
    max_runs.append(maximum_run)
max_runs
Out[41]:
[230, 240, 211, 246, 232, 222, 263, 231, 235, 248, 250, 241]
In [43]:
plt.style.use('ggplot')
fig=plt.gcf()
fig.set_size_inches(13,7.5)
plt.xticks(fontsize=15)
plt.yticks(fontsize=15)
plt.title("Highest team score in each season",fontsize=20)
ax=sns.barplot(x = season, y = max_runs, hue = teamid, dodge=False)
ax.legend(loc = 'center left', bbox_to_anchor=(1,0.7))

# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/teams_200s'
  plt.savefig(filename, bbox_inches = 'tight')

win percentage¶

In [44]:
matches_copy = pd.DataFrame
matches_copy = mat.copy(deep=True)

matches_copy['team1']=matches_copy['team1'].map(Teams)
matches_copy['team2']=matches_copy['team2'].map(Teams)
In [45]:
teams=(matches_copy['team1'].value_counts()+matches_copy['team2'].value_counts()).reset_index()
teams.columns=['team_name','Matches_played']
teams
Out[45]:
team_name Matches_played
0 CSK 164
1 DC 75
2 DD 177
3 GL 30
4 KKR 178
5 KTK 14
6 KXIP 176
7 MI 187
8 PW 46
9 RCB 180
10 RPS 30
11 RR 147
12 SRH 108
In [46]:
matches_copy['winner']=matches_copy['winner'].map(Teams)
In [47]:
wins=matches_copy['winner'].value_counts().reset_index()
wins.columns=['team_name','wins']
wins
Out[47]:
team_name wins
0 MI 109
1 CSK 100
2 KKR 92
3 RCB 84
4 KXIP 82
5 DD 77
6 RR 75
7 SRH 58
8 DC 29
9 RPS 15
10 GL 13
11 PW 12
12 KTK 6
In [48]:
player=teams.merge(wins,left_on='team_name',right_on='team_name',how='inner')
player['%win']=(player['wins']/player['Matches_played'])*100
player = player.sort_values('%win', ascending=False)
player
Out[48]:
team_name Matches_played wins %win
0 CSK 164 100 60.975610
7 MI 187 109 58.288770
12 SRH 108 58 53.703704
4 KKR 178 92 51.685393
11 RR 147 75 51.020408
10 RPS 30 15 50.000000
9 RCB 180 84 46.666667
6 KXIP 176 82 46.590909
2 DD 177 77 43.502825
3 GL 30 13 43.333333
5 KTK 14 6 42.857143
1 DC 75 29 38.666667
8 PW 46 12 26.086957
In [49]:
trace1=go.Bar(x=player['team_name'], y=player['Matches_played'], name='Total Matches')
trace2=go.Bar(x=player['team_name'], y=player['wins'], name='Matches wins')
In [50]:
matches_data=[trace1,trace2]
fig = py.iplot(matches_data)
In [53]:
def team1_vs_team2(team1,team2):
    mt1=mat[((mat['team1']==team1)|(mat['team2']==team1))&((mat['team1']==team2)|(mat['team2']==team2))]
    plt.style.use('ggplot')
    fig=plt.gcf()
    fig.set_size_inches(13,7.5)
    plt.xticks(fontsize=15)
    plt.yticks(fontsize=15)
    plt.title("ONE ON ONE CLASH",fontsize=20)
    ax=sns.countplot(x = 'season', hue='winner',data=mt1, palette='YlGnBu')
    ax.legend(loc = 'center left', bbox_to_anchor=(1,0.7))

    # save file
    if save_files:
      if not os.path.exists('plots'):
        os.makedirs('plots')
      filename = 'plots/CSKvsMI_EverySeason'
      plt.savefig(filename, bbox_inches = 'tight')
In [54]:
team1_vs_team2('Chennai Super Kings','Mumbai Indians')
In [55]:
data = matches_copy
micsk=data[np.logical_or(np.logical_and(data['team1']=='MI',data['team2']=='CSK'),np.logical_and(data['team2']=='MI',data['team1']=='CSK'))]
In [56]:
sns.set(style='dark')
fig=plt.gcf()
fig.set_size_inches(10,8)
sns.countplot(micsk['winner'],order=micsk['winner'].value_counts().index)
plt.text(-0.1,15,str(micsk['winner'].value_counts()['MI']),size=29,color='white')
plt.text(0.9,9,str(micsk['winner'].value_counts()['CSK']),size=29,color='white')
plt.xlabel('Winner',fontsize=15)
plt.ylabel('Count',fontsize=15)
plt.yticks(fontsize=0)
plt.title('MI vs CSK - head to head')
plt.show()

# save file
if save_files:
    if not os.path.exists('plots'):
        os.makedirs('plots')
    filename = 'plots/CSKvsMI_HeadToHead'
    fig.savefig(filename, bbox_inches='tight')

player analysis¶

In [57]:
### Players with most Man of the Match awards in IPL
mom = pd.DataFrame()
mom['Awards']=mat['player_of_match'].value_counts()
mom['Player'] =mom.index
mom=mom[:20]
plt.style.use('ggplot')
fig=plt.gcf()
fig.set_size_inches(18.5,8.5)
plt.xticks(rotation=90,fontsize=15)
plt.yticks(fontsize=16)
plt.title("Most Man Of The Matches in IPL (All Seasons)",fontsize=20)
ax=sns.barplot(x='Player',y='Awards', data=mom)
count=0
for p in ax.patches:
    height = p.get_height()
    ax.text(p.get_x()+p.get_width()/2.,height + 0.3,mom['Awards'].iloc[count],ha="center") 
    count+=1
    
# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/most_MOM'
  plt.savefig(filename, bbox_inches = 'tight')
In [58]:
dic=defaultdict(list)
for i in range(0,len(mat)):
    dic[mat.season.iloc[i]].append(mat.player_of_match.iloc[i])
player=[]
match=[]
year=[]
for i in sorted(dic.keys()):
    ctr=collections.Counter(dic[i])
    d={k: v for k, v in sorted(ctr.items(), key=lambda item: item[1],reverse=True)}
    player.append(list(d.keys())[0])
    match.append(list(d.values())[0])
    year.append(i)
plt.style.use('ggplot')
fig=plt.gcf()
fig.set_size_inches(18.5,7.5)
plt.xticks(fontsize=15)
plt.yticks(fontsize=15)
plt.title("Most Man Of The Match awards won by players each season", fontsize=20)
ax=sns.barplot(year, match)
count=0
for p in ax.patches:
    height = p.get_height()
    ax.text(p.get_x()+p.get_width()/2.,height + 0.15,player[count],ha="center") 
    count+=1
    
# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/most_MOM_EachSeason'
  plt.savefig(filename, bbox_inches = 'tight')
In [59]:
### TOP RUN GETTERS IN IPL HISTORY ###

top_runGetters = pd.DataFrame(deli.groupby('batsman')['batsman_runs'].sum().sort_values(ascending=False).head(10))
top_runGetters.reset_index(inplace=True)
top_runGetters.columns=['Batsman', 'Total runs']
plt.style.use('ggplot')
fig=plt.gcf()
fig.set_size_inches(18.5,7.5)

ax=sns.barplot(x='Batsman',y='Total runs', data=top_runGetters)

plt.title("Top Run Getters in IPL", fontsize=20, fontweight = 'bold')
plt.xlabel("Batsmen", size = 25)
plt.ylabel("Total Runs Scored", size=25)
plt.xticks(fontsize=15)
plt.yticks(fontsize=15)

# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/TopRunScorers'
  plt.savefig(filename, bbox_inches = 'tight')
In [60]:
top_runGetters.head()
Out[60]:
Batsman Total runs
0 V Kohli 5434
1 SK Raina 5415
2 RG Sharma 4914
3 DA Warner 4741
4 S Dhawan 4632
In [62]:
balls=deli.groupby('batsman')['ball'].count().reset_index()
balls
Out[62]:
batsman ball
0 A Ashish Reddy 196
1 A Chandila 7
2 A Chopra 75
3 A Choudhary 20
4 A Dananjaya 5
... ... ...
511 YV Takawale 183
512 Yashpal Singh 67
513 Younis Khan 7
514 Yuvraj Singh 2207
515 Z Khan 141

516 rows × 2 columns

In [63]:
runs=deli.groupby('batsman')['batsman_runs'].sum().reset_index()
runs
Out[63]:
batsman batsman_runs
0 A Ashish Reddy 280
1 A Chandila 4
2 A Chopra 53
3 A Choudhary 25
4 A Dananjaya 4
... ... ...
511 YV Takawale 192
512 Yashpal Singh 47
513 Younis Khan 3
514 Yuvraj Singh 2765
515 Z Khan 117

516 rows × 2 columns

In [64]:
### FOURS HIT BY BATSMEN IN IPL SO FAR ###

four=deli[deli['batsman_runs']==4]
#four
runs_4=four.groupby('batsman')['batsman_runs'].count().reset_index()
runs_4.columns=['batsman','4s']
runs_4
Out[64]:
batsman 4s
0 A Ashish Reddy 16
1 A Chopra 7
2 A Choudhary 1
3 A Flintoff 5
4 A Hales 13
... ... ...
405 YK Pathan 264
406 YV Takawale 26
407 Yashpal Singh 5
408 Yuvraj Singh 218
409 Z Khan 11

410 rows × 2 columns

In [65]:
### SIXES HIT BY BATSMEN IN IPL SO FAR ###

six=deli[deli['batsman_runs']==6]
#six

runs_6=six.groupby('batsman')['batsman_runs'].count().reset_index()
runs_6.columns=['batsman','6s']
runs_6
Out[65]:
batsman 6s
0 A Ashish Reddy 15
1 A Choudhary 1
2 A Flintoff 2
3 A Hales 6
4 A Mishra 5
... ... ...
331 Y Venugopal Rao 37
332 YK Pathan 161
333 YV Takawale 3
334 Yuvraj Singh 149
335 Z Khan 2

336 rows × 2 columns

In [66]:
player=pd.concat([runs,balls.iloc[:,1],runs_4.iloc[:,1],runs_6.iloc[:,1]],axis=1)
player

player.fillna(0,inplace=True)
player
Out[66]:
batsman batsman_runs ball 4s 6s
0 A Ashish Reddy 280 196 16.0 15.0
1 A Chandila 4 7 7.0 1.0
2 A Chopra 53 75 1.0 2.0
3 A Choudhary 25 20 5.0 6.0
4 A Dananjaya 4 5 13.0 5.0
... ... ... ... ... ...
511 YV Takawale 192 183 0.0 0.0
512 Yashpal Singh 47 67 0.0 0.0
513 Younis Khan 3 7 0.0 0.0
514 Yuvraj Singh 2765 2207 0.0 0.0
515 Z Khan 117 141 0.0 0.0

516 rows × 5 columns

In [67]:
player['strike_rate']=(player['batsman_runs']/player['ball'])*100
player
Out[67]:
batsman batsman_runs ball 4s 6s strike_rate
0 A Ashish Reddy 280 196 16.0 15.0 142.857143
1 A Chandila 4 7 7.0 1.0 57.142857
2 A Chopra 53 75 1.0 2.0 70.666667
3 A Choudhary 25 20 5.0 6.0 125.000000
4 A Dananjaya 4 5 13.0 5.0 80.000000
... ... ... ... ... ... ...
511 YV Takawale 192 183 0.0 0.0 104.918033
512 Yashpal Singh 47 67 0.0 0.0 70.149254
513 Younis Khan 3 7 0.0 0.0 42.857143
514 Yuvraj Singh 2765 2207 0.0 0.0 125.283190
515 Z Khan 117 141 0.0 0.0 82.978723

516 rows × 6 columns

In [68]:
grp=deli.groupby(['match_id','batsman','batting_team'])['batsman_runs'].sum().reset_index()
grp
Out[68]:
match_id batsman batting_team batsman_runs
0 1 A Choudhary RCB 6
1 1 BCJ Cutting SRH 16
2 1 CH Gayle RCB 32
3 1 DA Warner SRH 14
4 1 DJ Hooda SRH 16
... ... ... ... ...
11289 11415 RD Chahar MI 0
11290 11415 RG Sharma MI 15
11291 11415 SK Raina CSK 9
11292 11415 SN Thakur CSK 2
11293 11415 SR Watson CSK 84

11294 rows × 4 columns

In [69]:
maximum=grp.groupby('batsman')['batsman_runs'].max().reset_index()
maximum.columns=['batsman','max_runs']
maximum
Out[69]:
batsman max_runs
0 A Ashish Reddy 36
1 A Chandila 4
2 A Chopra 24
3 A Choudhary 15
4 A Dananjaya 4
... ... ...
511 YV Takawale 45
512 Yashpal Singh 20
513 Younis Khan 3
514 Yuvraj Singh 83
515 Z Khan 23

516 rows × 2 columns

In [70]:
player2=pd.concat([player,maximum.iloc[:,1]],axis=1)
player2

player2_df = pd.DataFrame(player2)

player2.fillna(0,inplace=True)
player2
Out[70]:
batsman batsman_runs ball 4s 6s strike_rate max_runs
0 A Ashish Reddy 280 196 16.0 15.0 142.857143 36
1 A Chandila 4 7 7.0 1.0 57.142857 4
2 A Chopra 53 75 1.0 2.0 70.666667 24
3 A Choudhary 25 20 5.0 6.0 125.000000 15
4 A Dananjaya 4 5 13.0 5.0 80.000000 4
... ... ... ... ... ... ... ...
511 YV Takawale 192 183 0.0 0.0 104.918033 45
512 Yashpal Singh 47 67 0.0 0.0 70.149254 20
513 Younis Khan 3 7 0.0 0.0 42.857143 3
514 Yuvraj Singh 2765 2207 0.0 0.0 125.283190 83
515 Z Khan 117 141 0.0 0.0 82.978723 23

516 rows × 7 columns

highest individual score¶

In [71]:
deli.groupby(['match_id','batsman','batting_team'])['batsman_runs'].sum().reset_index().sort_values(by='batsman_runs',ascending=False).head(10)
Out[71]:
match_id batsman batting_team batsman_runs
6200 411 CH Gayle RCB 175
900 60 BB McCullum KKR 158
8426 562 AB de Villiers RCB 133
10149 7935 RR Pant DD 130
9257 620 AB de Villiers RCB 129
5585 372 CH Gayle RCB 128
3135 206 M Vijay CSK 127
529 36 DA Warner SRH 126
7752 516 V Sehwag KXIP 122
10417 7953 SR Watson CSK 121

bowler analysis¶

In [72]:
deli['dismissal_kind'].unique()
Out[72]:
array([nan, 'caught', 'bowled', 'run out', 'lbw', 'caught and bowled',
       'stumped', 'retired hurt', 'hit wicket', 'obstructing the field'],
      dtype=object)

highest wickets¶

In [73]:
dismissal_kinds = ['caught', 'bowled', 'lbw', 'caught and bowled', 'stumped', 'hit wicket']
In [75]:
bowlers = deli[deli['dismissal_kind'].isin(dismissal_kinds)]
bowlers = deli.groupby('bowler').apply(lambda x: x['dismissal_kind'].dropna().reset_index(name='wickets'))
bowlers
Out[75]:
index wickets
bowler
A Ashish Reddy 0 80497 bowled
1 80502 caught
2 81750 caught
3 82253 caught
4 83084 caught
... ... ... ...
Z Khan 114 142480 caught
115 143476 bowled
116 144752 caught
117 147288 caught
118 149389 caught

8834 rows × 2 columns

In [76]:
bowlers_df = bowlers.groupby('bowler').count().reset_index()
top_bowlers= bowlers_df.sort_values(by='wickets', ascending=False)
top_bowlers_head=top_bowlers[top_bowlers.wickets>=40].head(10)
In [77]:
top_bowlers_head
Out[77]:
bowler index wickets
294 SL Malinga 188 188
84 DJ Bravo 168 168
6 A Mishra 165 165
113 Harbhajan Singh 161 161
231 PP Chawla 156 156
48 B Kumar 141 141
236 R Ashwin 138 138
300 SP Narine 137 137
330 UT Yadav 136 136
247 R Vinay Kumar 127 127
In [78]:
fig=plt.gcf()
fig.set_size_inches(15.5,5.5)
plt.xticks(fontsize=15, rotation=45)
plt.yticks(fontsize=15)
sns.barplot(top_bowlers_head['bowler'],top_bowlers['wickets'])
plt.title("Top Wicket Takers in IPL", fontsize=20, fontweight = 'bold')
plt.xlabel("Bowler", size = 25)
plt.ylabel("Wickets", size=25)

# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/TopWicketTakers'
  plt.savefig(filename, bbox_inches = 'tight')
In [80]:
batsmen = deli.groupby('batsman').apply(lambda x:np.sum(x['batsman_runs'])).reset_index(name="runs")
top_batsmen = batsmen.sort_values(by='runs', ascending=False)
top_batsmen=top_batsmen[top_batsmen.runs>=1000]
In [81]:
top_batsmen.rename(columns = {'batsman': 'player'}, inplace = True)
top_batsmen.head()
Out[81]:
player runs
486 V Kohli 5434
428 SK Raina 5415
367 RG Sharma 4914
112 DA Warner 4741
392 S Dhawan 4632
In [82]:
top_bowlers.rename(columns = {'bowler': 'player'}, inplace = True)
In [89]:
#Adding some more Feature Columns to deliveries Dataset
dic = dict()
for match_id in mat['id'].unique():
    dic[match_id] = mat[mat['id']==match_id]['season'].values[0]
deli['season'] = [dic[i] for i in deli['match_id']]
In [90]:
# Selecting only the Death Overs i.e. 16-20 overs
deli = deli[deli['over']>=16]
deli.shape
Out[90]:
(40400, 22)
In [91]:
mat.head(1)
Out[91]:
id season city date team1 team2 toss_winner toss_decision result dl_applied winner win_by_runs win_by_wickets player_of_match venue umpire1 umpire2 umpire3
0 1 2017 Hyderabad 2017-04-05 Sunrisers Hyderabad Royal Challengers Bangalore Royal Challengers Bangalore field normal 0 Sunrisers Hyderabad 35 0 Yuvraj Singh Rajiv Gandhi International Stadium, Uppal AY Dandekar NJ Llong NaN
In [92]:
deli.head(1)
Out[92]:
match_id inning batting_team bowling_team over ball batsman non_striker bowler is_super_over ... legbye_runs noball_runs penalty_runs batsman_runs extra_runs total_runs player_dismissed dismissal_kind fielder season
93 1 1 SRH RCB 16 1 MC Henriques Yuvraj Singh YS Chahal 0 ... 0 0 0 0 0 0 NaN NaN NaN 2017

1 rows × 22 columns

venue analysis¶

In [104]:
sta= pd.DataFrame()
sta['Count']=mat['venue'].value_counts()
sta['Venue'] =sta.index
sta=sta[:20]
plt.style.use('ggplot')
fig=plt.gcf()
fig.set_size_inches(18.5,10.5)
plt.ylabel("Venue", size=20)
plt.xticks(fontsize=20)
plt.yticks(fontsize=16)
plt.title("Matches Played in Different Stadiums",fontsize=20)
ax=sns.barplot(x='Count', y='Venue', data=sta)
count=0

for i, v in enumerate(sta['Count']):
    ax.text(v + 1.5, i, str(v),va="center", fontdict=dict(fontsize=20))

# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/matches_In_Diff_Stadiums'
  plt.savefig(filename, bbox_inches = 'tight')
In [98]:
df = pd.DataFrame()
df1 = pd.DataFrame()
chunks = []
chunks.append(mat['toss_winner'][mat['toss_decision']=='bat'])
chunks.append(mat['team2'][(mat['toss_decision']=='field') & (mat['toss_winner'] == mat['team1'])])
chunks.append(mat['team1'][(mat['toss_decision']=='field') & (mat['toss_winner'] == mat['team2'])])

df = pd.DataFrame(pd.concat(chunks))
df = df.sort_index()
df.columns = ['team']

indexes = df[(df['team'] == mat['winner'])&(df.index == mat.index)].index

df1['venue'] = mat['venue'].value_counts()
df1['win_count'] = mat[mat.index.isin(indexes)]['venue'].value_counts()
df1['win %'] = 100 * df1['win_count']/df1['venue'] 
df1 = df1.sort_values(by = ['win %'],ascending = False)
df1 = df1[:20]
winPercent = df1['win %'].astype('int')
plt.style.use('ggplot')
fig=plt.gcf()
fig.set_size_inches(18.5,10.5)
plt.xticks(fontsize=16)
plt.yticks(fontsize=16)
plt.title("Percent of matches won batting first in Different Stadiums(Top 20)",fontsize=19)
ax=sns.barplot(winPercent[:20], df1[:20].index)
plt.xlabel("Win % batting first", size=20)
count=0

for i, v in enumerate(winPercent):
    ax.text(v + 1.5, i, str(v),va="center", fontdict=dict(fontsize=20))

# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/percent_BatFirstWin_StadiumWise'
  plt.savefig(filename, bbox_inches = 'tight')
In [ ]:
In [99]:
df1 = pd.DataFrame()
 
df1['venue'] = mat['venue'].value_counts()
df1['win_count'] = mat[mat.index.isin(indexes)]['venue'].value_counts()
df1['win %'] = 100 * df1['win_count']/df1['venue'] 
df1 = df1.sort_values(by = ['win %'])
df1['win % balling first'] = 100 - df1['win %']
df1 = df1[:20]
df1['win % balling first'] = df1['win % balling first'].astype('int')
plt.style.use('ggplot')
fig=plt.gcf()
fig.set_size_inches(18.5,10.5)
plt.xticks(fontsize=16)
plt.yticks(fontsize=16)
plt.title("Percent of matches won fielding first in Different Stadiums(Top 20)",fontsize=20)
ax=sns.barplot(df1['win % balling first'],df1.index)
plt.xlabel("Win % fielding first", size=20)
count=0

for i, v in enumerate(df1['win % balling first']):
    ax.text(v + 1.5, i, str(v),va="center", fontdict=dict(fontsize=20))


# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/percent_FieldFirstWin_StadiumWise'
  plt.savefig(filename, bbox_inches = 'tight')

toss analysis¶

In [100]:
temp_series = mat.toss_decision.value_counts()
labels = (np.array(temp_series.index))
sizes = (np.array((temp_series / temp_series.sum())*100))
colors = ['gold', 'lightskyblue']
fig = plt.figure(figsize=(5,5))
plt.pie(sizes, labels=labels, colors=colors, autopct='%1.1f%%', shadow=True, startangle=90)
plt.title("Toss decision percentage")
plt.show()

# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/Toss_Decision_Percent_pie'
  fig.savefig(filename, bbox_inches = 'tight')
In [102]:
num_of_wins = (mat.win_by_wickets>0).sum()
num_of_loss = (mat.win_by_wickets==0).sum()
labels = ["Wins", "Loss"]
total = float(num_of_wins + num_of_loss)
sizes = [(num_of_wins/total)*100, (num_of_loss/total)*100]
colors = ['gold', 'lightskyblue']
fig = plt.figure(figsize=(5,5))
plt.pie(sizes, labels=labels, colors=colors, autopct='%1.1f%%', shadow=True, startangle=90)
plt.title("Win percentage batting second")
plt.show()

# save file
if save_files:
  if not os.path.exists('plots'):
    os.makedirs('plots')
  filename = 'plots/WinPercent_Chasing'
  fig.savefig(filename, bbox_inches = 'tight')
In [ ]: